Danish ul Khairi

Mr. Danish ul Khairi

orcid

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
31 January 2023 in Future Transportation
Verified
Journal Article
Open access
PDF file
DKDanish ul Khairi
Danish ul Khairi
FAFerheen Ayaz
Ferheen Ayaz
Nagham SaeedNagham Saeed
Nagham Saeed
KAKamran Ahsan
Kamran Ahsan
SASyed Zeeshan Ali
Syed Zeeshan Ali
Intelligent transportation systems (ITS) is a broad area that encompasses vehicle identification, classification, monitoring, surveillance, prediction, management, reduction of traffic jams, license plate recognition, etc. Machine learning has practical and significant applications in ITS. Intelligent transportation systems rely heavily on vehicle classification for traffic management and monitoring. This research uses convolutional neural networks to classify cars at fine-grained classifications (make and model). Numerous obstacles must be overcome in order to complete the task, the greatest of which are intra- and inter-class similarities between the manufacturer and model of vehicles, different lighting effects, the shape and size of the vehicle, shadows, camera view angle, background, vehicle speed, colour occlusion and environmental conditions. This paper studies various machine learning algorithms used for the fine-grained classification of vehicles and presents a comparative analysis in terms of accuracy and the size of the implemented deep convolutional neural network (DCNN). Specifically, four DCNN models, mobilenet-v2, inception-v3, vgg-19 and resnet-50, are evaluated with three datasets, BMW-10, Stanford Cars and PAKCars. The evaluation results show that mobileNet-v2 is the smallest model as it is not computationally intensive due to depthwise separable convolution. However, resnet-50 and vgg-19 outperform inception-v3 and mobilenet-v2 in terms of accuracy due to their complex structure.

10 January 2024 in Research Square
Verified
Preprint
Open access
PDF file
SASyed Zeeshan Ali
Syed Zeeshan Ali
KAKamran Ahsan
Kamran Ahsan
DKDanish ul Khairi
Danish ul Khairi
In this study, the free space approach is utilized to calculate the relative per-mittivity of FR4 by utilizing the Nicholson-Ross-Weir Conversion. By examining the scattering characteristics, the free space technique offers a practical tool for describing dielectric materials. The simulations were run on CST-2019, and the frequency range of 8.5 GHz to 11.5 GHz was chosen. Experimental measurements were carried out utilizing a Vector Network Analyzer, To further reduce outside influences and assure accurate measurements in a controlled setting, an anechoic chamber was used. The outcomes of the simulations and actual measurements show the significance of the Nicholson-Ross-Weir Conversion and free space approach in calculating the relative permittivity of FR4. The correctness and dependability of the suggested technique are confirmed by the good agreement between the simulated and measured outcomes. This study makes a contribution to the field of electromagnetic characterization and offers a useful method for figuring out FR4’s dielectric characteristics. The results of this study have substantial effects on PCB design and optimization as well as other high-frequency electronic devices that operate in the frequency band under consideration.

8 June 2023 in Research Square
Verified
Preprint
Open access
PDF file
DKDanish ul Khairi
Danish ul Khairi
KAKamran Ahsan
Kamran Ahsan
GBGran Badshah
Gran Badshah
SASyed Zeeshan Ali
Syed Zeeshan Ali
SRSyed Akhter Raza
Syed Akhter Raza
MSMuhammad Shiraz
Muhammad Shiraz
This study presents a comparative analysis of classical and deep learning approaches for the classification of apple fruit quality, within the broader context of machine vision applications. Emphasizing the importance of the fruit's physical appearance in meeting market standards, the research explores the performance of classical methods such as Support Vector Machine (SVM), K-nearest neighbour (KNN), and Decision Tree (DT), in comparison to deep learning methods like Mobile-Net and Convolutional Neural Network (CNN) with self-design. A self-created dataset comprising 150 apple fruit images categorized into fresh, mid, and rotten classes are utilized, with an 80:20 training-test split. The evaluation of the approaches reveals promising results, with SVM achieving 86% accuracy, DT achieving 93%, KNN achieving 96.6%, and Mobile-Net and CNN achieving 97% accuracy. Notably, the study demonstrates the efficiency of these methods in classifying different quality classes of apple fruit. This research contributes to the existing knowledge in agricultural quality control and provides insights for researchers seeking suitable algorithms in the field of machine vision for fruit quality assessment, ultimately advancing the application of machine vision techniques in real-world scenarios.

10 December 2025 in PLOS ONE
Verified
Journal Article
Open access
DKDanish ul Khairi
Danish ul Khairi
KAKamran Ahsan
Kamran Ahsan
SASyed Zeeshan Ali
Syed Zeeshan Ali
WAWadee ALHALABI
Wadee ALHALABI
SASomayah Albaradei
Somayah Albaradei
MAMuhammad Shahid Anwar
Muhammad Shahid Anwar
This study investigates hybrid quantum neural networks for fruit quality assessment, with a focus on the impact of the entangling gate choice. Two architectures were developed: NNQEv1, utilizing controlled-NOT (CNOT) gates, and NNQEv2, employing controlled-phase (CZ) gates. A theoretical justification is provided, based on gate decomposition and hardware-aware noise considerations, suggesting the CZ-based architecture is likely to be more stable. The performance of the models was evaluated through the computational execution of their quantum circuits on classical hardware and compared against classical and state-of-the-art deep learning models. The proposed models demonstrated competitive performance, achieving test accuracies of 98.7% on MNIST, 98.6% on the FruitQ dataset, and 96.7% on a custom, data-scarce Apple dataset. The experimental results align with the theoretical analysis: the CZ-based NNQEv2 model, when compared to the CNOT-based NNQEv1, consistently showed more stable training dynamics and yielded tighter confidence intervals in cross-validation. This work presents a foundational, computational study on the role of gate-level design choices, intended to inform the development of future quantum machine learning algorithms.

9 February 2024 in Research Square
Verified
Preprint
Open access
PDF file
SASyed Zeeshan Ali
Syed Zeeshan Ali
KAKamran Ahsan
Kamran Ahsan
DKDanish ul Khairi
Danish ul Khairi
SRSyed Akhter Raza
Syed Akhter Raza
This study presents the design and development of a truncated spherical dielec-tric lens antenna. The objective was to enhance and optimize the directivity and beamwidth of the feed antenna by adjusting the feed position around the focal point of the lens. Polyetrafluoroethylene (PTFE), with a dielectric constant of 2.1, was chosen as the lens medium. Simulations were conducted using CST Studio Suite over a frequency range of 8 to 12 GHz. Experimental validation was performed using a vector network analyzer (VNA) and an anechoic chamber. The optimized dimensions of the dielectric lens antenna were determined as follows: lens thickness (T) = 24.34 mm, center of curvature (R) = 55 mm, and lens aperture (L) = 110 mm. This configuration achieved a directivity of 21.9 dBi and a beamwidth of 13.1°. The resulting dielectric lens antenna exhibits a compact size, simple geometric shape, and ease of design and construction. These characteristics make it highly suitable for RF applications, particularly when energy concentration is required.